Hindawi Publishing Corporation Boundary Value Problems Volume 2011, Article ID 715836, 12 pages doi:10.1155/2011/715836 Research Article Global Structure of Nodal Solutions for Second-Order m-Point Boundary Value Problems with Superlinear Nonlinearities Yulian An Department of Mathematics, Shanghai Institute of Technology, Shanghai 200235, China Correspondence should be addressed to Yulian An, an yulian@tom.com Received 8 May 2010; Revised 1 August 2010; Accepted 23 September 2010 Academic Editor: Feliz Manuel Minhós Copyright q 2011 Yulian An. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. We consider the nonlinear eigenvalue problems u λfu 0, 0 < t < 1, u0 0, u1 m−2 m−2 i1 αi uηi , where m ≥ 3, ηi ∈ 0, 1, and αi > 0 for i 1, . . . , m − 2, with i1 αi < 1, and 1 f ∈ C R\{0}, R ∩ CR, R satisfies fss > 0 for s / 0, and f0 ∞, where f0 lim|s| → 0 fs/s. We investigate the global structure of nodal solutions by using the Rabinowitz’s global bifurcation theorem. 1. Introduction We study the global structure of nodal solutions of the problem u λfu 0, u0 0, u1 t ∈ 0, 1, m−2 αi u ηi . 1.1 1.2 i1 Here m ≥ 3, ηi ∈ 0, 1, and αi > 0 for i 1, . . . , m − 2 with m−2 i1 αi < 1; λ is a positive parameter, and f ∈ C1 R \ {0}, R ∩ CR, R. In the case that f0 ∈ 0, ∞, the global structure of nodal solutions of nonlinear secondorder m-point eigenvalue problems 1.1, 1.2 have been extensively studied; see 1–5 and the references therein. However, relatively little is known about the global structure of solutions in the case that f0 ∞, and few global results were found in the available literature when f0 ∞ f∞ . The likely reason is that the global bifurcation techniques cannot be 2 Boundary Value Problems used directly in the case. On the other hand, when m-point boundary value condition 1.2 is concerned, the discussion is more difficult since the problem is nonsymmetric and the corresponding operator is disconjugate. In 6, we discussed the global structure of positive solutions of 1.1, 1.2 with f0 ∞. However, to the best of our knowledge, there is no paper to discuss the global structure of nodal solutions of 1.1, 1.2 with f0 ∞. In this paper, we obtain a complete description of the global structure of nodal solutions of 1.1, 1.2 under the following assumptions: A1 αi > 0 for i 1, . . . , m − 2, with 0 < m−2 i1 αi < 1; A2 f ∈ C1 R \ {0}, R ∩ CR, R satisfies fss > 0 for s / 0; A3 f0 : lim|s| → 0 fs/s ∞; A4 f∞ : lim|s| → ∞ fs/s ∈ 0, ∞. Let Y C0, 1 with the norm u∞ max |ut|. 1.3 t∈0,1 Let X u ∈ C 0, 1 | u0 0, u1 1 i1 E αi u ηi , m−2 u ∈ C2 0, 1 | u0 0, u1 m−2 αi u ηi 1.4 i1 with the norm uX max u∞ , u ∞ , u max u∞ , u ∞ , u ∞ , 1.5 respectively. Define L : E → Y by setting Lu : −u , u ∈ E. 1.6 Then L has a bounded inverse L−1 : Y → E and the restriction of L−1 to X, that is, L−1 : X → X is a compact and continuous operator; see 1, 2, 6. 0. For For any C1 function u, if ux0 0, then x0 is a simple zero of u if u x0 / any integer k ≥ 1 and any ν ∈ {, −}, define sets Sνk , Tkν ⊂ C2 0, 1 consisting of functions u ∈ C2 0, 1 satisfying the following conditions: Sνk : i u0 0, νu 0 > 0, ii u has only simple zeros in 0, 1 and has exactly k − 1 zeros in 0, 1; Tkν : i u0 0, νu 0 > 0 and u 1 / 0, ii u has only simple zeros in 0, 1 and has exactly k zeros in 0, 1, iii u has a zero strictly between each two consecutive zeros of u . Boundary Value Problems 3 Remark 1.1. Obviously, if u ∈ Tkν , then u ∈ Sνk or u ∈ Sνk1 . The sets Tkν are open in E and disjoint. Remark 1.2. The nodal properties of solutions of nonlinear Sturm-Liouville problems with separated boundary conditions are usually described in terms of sets similar to Sνk ; see 7. However, Rynne 1 stated that Tkν are more appropriate than Sνk when the multipoint boundary condition 1.2 is considered. Next, we consider the eigenvalues of the linear problem Lu λu, u ∈ E. 1.7 We call the set of eigenvalues of 1.7 the spectrum of L and denote it by σL. The following lemmas or similar results can be found in 1–3. Lemma 1.3. Let A1 hold. The spectrum σL consists of a strictly increasing positive sequence of eigenvalues λk , k 1, 2, . . . , with corresponding eigenfunctions ϕk x sin λk x. In addition, i limk → ∞ λk ∞; ii ϕk ∈ Tk , for each k ≥ 1, and ϕ1 is strictly positive on 0, 1. We can regard the inverse operator L−1 : Y → E as an operator L−1 : Y → Y. In this setting, each λk , k 1, 2, . . . , is a characteristic value of L−1 , with algebraic multiplicity −1 j defined to be dim ∞ j1 NI − λk L , where N denotes null-space and I is the identity on Y. Lemma 1.4. Let A1 hold. For each k ≥ 1, the algebraic multiplicity of the characteristic value λk , k 1, 2, . . . , of L−1 : Y → Y is equal to 1. Let E R × E under the product topology. As in 7, we add the points {λ, ∞ | λ ∈ R} to our space E. Let Φνk R × Tkν . Let Σνk denote the closure of set of those solutions of 1.1, 1.2 which belong to Φνk . The main results of this paper are the following. Theorem 1.5. Let (A1)–(A4) hold. a If f∞ 0, then there exists a subcontinuum Cνk of Σνk with 0, 0 ∈ Cνk and ProjR Cνk 0, ∞. 1.8 b If f∞ ∈ 0, ∞, then there exists a subcontinuum Cνk of Σνk with 0, 0 ∈ Cνk , ProjR Cνk ⊆ λ1 0, . f∞ 1.9 c If f∞ ∞, then there exists a subcontinuum Cνk of Σνk with 0, 0 ∈ Cνk , ProjR Cνk is a bounded closed interval, and Cνk approaches 0, ∞ as u → ∞. 4 Boundary Value Problems Theorem 1.6. Let (A1)–(A4) hold. a If f∞ 0, then 1.1, 1.2 has at least one solution in Tkν for any λ ∈ 0, ∞. b If f∞ ∈ 0, ∞, then 1.1, 1.2 has at least one solution in Tkν for any λ ∈ 0, λ1 /f∞ . c If f∞ ∞, then there exists λ∗ > 0 such that 1.1, 1.2 has at least two solutions in Tkν for any λ ∈ 0, λ∗ . We will develop a bifurcation approach to treat the case f0 ∞. Crucial to this approach is to construct a sequence of functions {f n } which is asymptotic linear at 0 and satisfies f n −→ f, f n 0 −→ ∞. 1.10 By means of the corresponding auxiliary equations, we obtain a sequence of unbounded νn components {Ck } via Rabinowitz’s global bifurcation theorem 8, and this enables us to find unbounded components Cνk satisfying νn 0, 0 ∈ Cνk ⊂ lim sup Ck . 1.11 The rest of the paper is organized as follows. Section 2 contains some preliminary propositions. In Section 3, we use the global bifurcation theorems to analyse the global behavior of the components of nodal solutions of 1.1, 1.2. 2. Preliminaries Definition 2.1 see 9. Let W be a Banach space and {Cn | n 1, 2, . . .} a family of subsets of W. Then the superior limit D of {Cn } is defined by D : lim sup Cn {x ∈ W | ∃{ni } ⊂ N and xni ∈ Cni , such that xni −→ x}. n→∞ 2.1 Lemma 2.2 see 9. Each connected subset of metric space W is contained in a component, and each connected component of W is closed. Lemma 2.3 see 6. Assume that i there exist zn ∈ Cn n 1, 2, . . . and z∗ ∈ W, such that zn → z∗ ; ii rn ∞, where rn sup{x | x ∈ Cn }; iii for all R > 0, ∞ n1 Cn ∩ BR is a relative compact set of W, where BR {x ∈ W | x ≤ R}. Then there exists an unbounded connected component C in D and z∗ ∈ C. 2.2 Boundary Value Problems 5 Define the map Tλ : Y → E by Tλ ut λ 1 2.3 Ht, sfusds, 0 where i1 αi G ηi , s Ht, s Gt, s t, 1 − m−2 i1 αi ηi m−2 Gt, s ⎧ ⎨1 − ts, 0 ≤ s ≤ t ≤ 1, ⎩t1 − s, 0 ≤ t ≤ s ≤ 1. 2.4 It is easy to verify that the following lemma holds. Lemma 2.4. Assume that (A1)-(A2) hold. Then Tλ : Y → E is completely continuous. For r > 0, let Ωr {u ∈ Y | u∞ < r}. 2.5 Lemma 2.5. Let (A1)-(A2) hold. If u ∈ ∂Ωr , r > 0, then r Tλ u∞ ≤ λM m−2 1 1− i1 αi m−2 i1 αi ηi 1 Gs, sds, 2.6 0 r 1 max0≤|s|≤r {|fs|}. where M Proof. The proof is similar to that of Lemma 3.5 in 6; we omit it. Lemma 2.6. Let (A1)-(A2) hold, and {μl , yl } ⊂ Φνk is a sequence of solutions of 1.1, 1.2. Assume that μl ≤ C0 for some constant C0 > 0, and liml → ∞ yl ∞. Then lim yl ∞ ∞. 2.7 l→∞ Proof. From the relation yl t 1 μl 0 Ht t, sfyl sds. Then y ≤ C0 1 l ∞ μl 1 0 Ht, sfyl sds, we conclude that y1 t m−2 1− i1 αi m−2 i1 αi ηi 1 f yl s ds, 0 which implies that {yl ∞ } is bounded whenever {yl ∞ } is bounded. 2.8 6 Boundary Value Problems 3. Proof of the Main Results For each n ∈ N, define f n s : R → R by f n s 1 1 , ∞ ∪ −∞, − , n n 1 1 s∈ − , . n n ⎧ ⎪ ⎪ ⎪fs, ⎨ s∈ ⎪ ⎪ ⎪ ⎩nf 1 s, n 3.1 Then f n ∈ CR, R ∩ C1 R \ {±1/n}, R with f n ss > 0, f n ∀s / 0, 0 nf 1 . n 3.2 By A3, it follows that lim f n ∞. n→∞ 0 3.3 Now let us consider the auxiliary family of the equations u λf n u 0, u0 0, u1 t ∈ 0, 1, m−2 αi u ηi . 3.4 3.5 i1 Lemma 3.1 see 1, Proposition 4.1. Let (A1), (A2) hold. If μ, u ∈ E is a nontrivial solution of 3.4, 3.5, then u ∈ Tkν for some k, ν. Let ζn ∈ CR, R be such that f n u f n 0 uζ n 1 u ζn u. u nf n 3.6 Note that ζn s 0. |s| → 0 s lim 3.7 Let us consider Lu − λ f n u λζn u 0 as a bifurcation problem from the trivial solution u ≡ 0. 3.8 Boundary Value Problems 7 Equation 3.8 can be converted to the equivalent equation ut 1 Ht, s λ f n us λζn us ds 0 0 : λL −1 f n 0 u· t λL−1 ζn u· t. 3.9 Further we note that L−1 ζn u ou for u near 0 in E. The results of Rabinowitz 8 for 3.8 can be stated as follows. For each integer k ≥ νn 1, ν ∈ {, −}, there exists a continuum {Ck } of solutions of 3.8 joining λk /f n 0 , 0 to νn infinity in E. Moreover, {Ck } \ {λk /f n 0 , 0} ⊂ Φνk . νn Proof of Theorem 1.5. Let us verify that {Ck Since } satisfies all of the conditions of Lemma 2.3. λk λk 0, lim n n → ∞ nf1/n f 0 lim n→∞ 3.10 condition i in Lemma 2.3 is satisfied with z∗ 0, 0. Obviously νn ∞, rn sup λ y | λ, y ∈ Ck 3.11 and accordingly, ii holds. iii can be deduced directly from the Arzela-Ascoli Theorem and νn the definition of f n . Therefore, the superior limit of {Ck }, Dkν , contains an unbounded ν ν connected component Ck with 0, 0 ∈ Ck . From the condition A2, applying Lemma 2.2 with p 2 in 10, we can show that the initial value problem v λfv 0, vt0 0, t ∈ 0, 1, vt0 β 3.12 has a unique solution on 0, 1 for every t0 ∈ 0, 1 and β ∈ R. Therefore, any nontrivial 0. Meanwhile, A1 implies solution u of 1.1, 1.2 has only simple zeros in 0, 1 and u 0 / νn ν that u 1 0 1, proposition 4.1. Since C ⊂ Φ , we conclude that Cνk ⊂ Φνk . Moreover, / k k ν ν Ck ⊂ Σk by 1.1 and 1.2. We divide the proof into three cases. Case 1 f∞ 0. In this case, we show that ProjR Cνk 0, ∞. Assume on the contrary that sup λ | λ, u ∈ Cνk < ∞, 3.13 8 Boundary Value Problems then there exists a sequence {μl , yl } ⊂ Cνk such that lim yl ∞, l→∞ μl ≤ C0 , 3.14 for some positive constant C0 depending not on l. From Lemma 2.6, we have lim yl ∞ ∞. 3.15 l→∞ Set vl t yl t/yl ∞ . Then vl ∞ 1. Now, choosing a subsequence and relabelling if necessary, it follows that there exists μ∗ , v∗ ∈ 0, C0 × E with v∗ ∞ 1, 3.16 such that lim μl , vl μ∗ , v∗ , l→∞ in R × E. 3.17 Since lim|u| → ∞ fu/u 0, we can show that f yl t 0. lim l → ∞ yl ∞ 3.18 The proof is similar to that of the step 1 of Theorem 1 in 7; we omit it. So, we obtain v∗ t μ∗ · 0 0, v∗ 0 0, v∗ 1 t ∈ 0, 1, m−2 αi v∗ ηi , 3.19 3.20 i1 and subsequently, v∗ t ≡ 0 for t ∈ 0, 1. This contradicts 3.16. Therefore sup λ | λ, y ∈ Cνk ∞. 3.21 Case 2 f∞ ∈ 0, ∞. In this case, we can show easily that C joins 0, 0 with λk /f∞ , ∞ by using the same method used to prove Theorem 5.1 in 2. Case 3 f∞ ∞. In this case, we show that Cνk joins 0, 0 with 0, ∞. Let {μl , yl } ⊂ Cνk be such that μl yl −→ ∞, l −→ ∞. 3.22 Boundary Value Problems 9 If {yl } is bounded, say, yl ≤ M1 , for some M1 depending not on l, then we may assume that lim μl ∞. 3.23 l→∞ Taking subsequences again if necessary, we still denote {μl , yl } such that {yl } ⊂ Tkν ∩ Sνk . If {yl } ⊂ Tkν ∩ Sνk1 , all the following proofs are similar. Let 0 τl0 < τl1 < · · · < τlk−1 3.24 j denote the zeros of yl in 0, 1. Then, after taking a subsequence if necessary, liml → ∞ τl : j 0 k τ∞ , j ∈ {0, 1, . . . , k − 1}. Clearly, τ∞ 0. Set τ∞ 1. We can choose at least one subinterval j j1 j τ∞ , τ∞ I∞ which is of length at least 1/k for some j ∈ {0, 1, . . . , k − 1}. Then, for this j1 j j j1 j j, τl − τl > 3/4k if l is large enough. Put τl , τl Il . j Obviously, for the above given k, ν and j, yl t have the same sign on Il for all l. Without loss of generality, we assume that t ∈ Il . j 3.25 max|ul t| ≤ M1 . 3.26 yl t > 0, Moreover, we have j t∈Il Combining this with the fact ! " f yl t fs | 0 < s ≤ M1 > 0, ≥ inf yl t s j j1 , t ∈ τl , τl 3.27 and using the relation yl t f yl t yl t 0, μl yl t j j j1 , t ∈ τl , τl 3.28 j1 we deduce that yl must change its sign on τl , τl if l is large enough. This is a contradiction. Hence {yl } is unbounded. From Lemma 2.6, we have that lim yl ∞ ∞. l→∞ 3.29 Note that {μl , yl } satisfies the autonomous equation yl μl f yl 0, t ∈ 0, 1. 3.30 10 Boundary Value Problems We see that yl consists of a sequence of positive and negative bumps, together with a truncated bump at the right end of the interval 0, 1, with the following properties ignoring the truncated bump see, 1: i all the positive resp., negative bumps have the same shape the shapes of the positive and negative bumps may be different; ii each bump contains a single zero of yl , and there is exactly one zero of yl between consecutive zeros of yl ; iii all the positive negative bumps attain the same maximum minimum value. Armed with this information on the shape of yl , it is easy to show that for the above j given Il , {yl I j ,∞ : maxI j yl t}∞ l1 is an unbounded sequence. That is l l lim yl I j ,∞ ∞. l→∞ 3.31 l j Since yl is concave on Il , for any σ > 0 small enough, yl t ≥ σ yl I j ,∞ , l j j1 ∀t ∈ τl σ, τl − σ . 3.32 j This together with 3.31 implies that there exist constants α, β with α, β ⊂ I∞ , such that lim yl t ∞, l→∞ # $ uniformly for t ∈ α, β . 3.33 Hence, we have f yl t ∞, lim l → ∞ yl t # $ uniformly for t ∈ α, β . 3.34 Now, we show that liml → ∞ μl 0. Suppose on the contrary that, choosing a subsequence and relabeling if necessary, μl ≥ b0 for some constant b0 > 0. This implies that f yl t lim μl ∞, l→∞ yl t # $ uniformly for t ∈ α, β . 3.35 From 3.28 we obtain that yl must change its sign on α, β if l is large enough. This is a contradiction. Therefore liml → ∞ μl 0. Proof of Theorem 1.6. a and b are immediate consequence of Theorem 1.5a and b, respectively. To prove c, we rewrite 1.1, 1.2 to uλ 1 0 Ht, sfusds Tλ ut. 3.36 Boundary Value Problems 11 By Lemma 2.5, for every r > 0 and u ∈ ∂Ωr , r Tλ u∞ ≤ λM 1 m−2 1− i1 αi m−2 i1 αi ηi 1 Gs, sds, 3.37 0 r 1 max0≤|s|≤r {|fs|}. where M Let λr > 0 be such that r λr M 1 m−2 1− i1 αi m−2 i1 αi ηi 1 Gs, sds r. 3.38 0 Then for λ ∈ 0, λr and u ∈ ∂Ωr , Tλ u∞ < u∞ . 3.39 Σνk ∩ {λ, u ∈ 0, ∞ × E | 0 < λ < λr , u ∈ E : u∞ r} ∅. 3.40 This means that By Lemma 2.6 and Theorem 1.5, it follows that Cνk is also an unbounded component joining 0, 0 and 0, ∞ in 0, ∞ × Y . Thus, 3.40 implies that for λ ∈ 0, λr , 1.1, 1.2 has at least two solutions in Tkν . Acknowledgments The author is very grateful to the anonymous referees for their valuable suggestions. This paper was supported by NSFC no.10671158, 11YZ225, YJ2009-16 no.A06/1020K096019. References 1 B. P. Rynne, “Spectral properties and nodal solutions for second-order, m-point, boundary value problems,” Nonlinear Analysis: Theory, Methods & Applications, vol. 67, no. 12, pp. 3318–3327, 2007. 2 R. Ma and D. O’Regan, “Nodal solutions for second-order m-point boundary value problems with nonlinearities across several eigenvalues,” Nonlinear Analysis: Theory, Methods & Applications, vol. 64, no. 7, pp. 1562–1577, 2006. 3 X. Xu, “Multiple sign-changing solutions for some m-point boundary-value problems,” Electronic Journal of Differential Equations, vol. 2004, no. 89, pp. 1–14, 2004. 4 S. Jingxian, X. Xian, and D. O’Regan, “Nodal solutions for m-point boundary value problems using bifurcation methods,” Nonlinear Analysis: Theory, Methods & Applications, vol. 68, no. 10, pp. 3034–3046, 2008. 5 Y. An and R. Ma, “Global behavior of the components for the second order m-point boundary value problems,” Boundary Value Problems, vol. 2008, Article ID 254593, 10 pages, 2008. 6 R. Ma and Y. An, “Global structure of positive solutions for superlinear second order m-point boundary value problems,” Topological Methods in Nonlinear Analysis, vol. 34, no. 2, pp. 279–290, 2009. 7 R. Ma and B. Thompson, “Nodal solutions for nonlinear eigenvalue problems,” Nonlinear Analysis: Theory, Methods & Applications, vol. 59, no. 5, pp. 707–718, 2004. 12 Boundary Value Problems 8 P. H. Rabinowitz, “Some global results for nonlinear eigenvalue problems,” Journal of Functional Analysis, vol. 7, pp. 487–513, 1971. 9 G. T. Whyburn, Topological Analysis, Princeton Mathematical Series. No. 23, Princeton University Press, Princeton, NJ, USA, 1958. 10 Y. Naito and S. Tanaka, “Sharp conditions for the existence of sign-changing solutions to equations involving the one-dimensional p-Laplacian,” Nonlinear Analysis: Theory, Methods & Applications, vol. 69, no. 9, pp. 3070–3083, 2008.